hamradiotools

Nexus 1.7.6

Released 21 August 2026. The current release is 1.24.0. Download it · Every version

Fixed

  • The Call Roster shows the full CQ, not just "CQ".

    A station calling CQ DX looked exactly like one calling a general CQ, so you could click to work him and only find out from the Band Activity pane that he wants DX and will ignore you. The roster now shows CQ DX, CQ POTA, CQ NA and the rest, in a colour that stands out while you are scanning, with a note on hover about what it means for answering.

  • A failed OmniRig connection stops telling you to check a serial port.

    If Test CAT could not read a frequency over OmniRig, the message advised checking the serial port, baud rate and CI-V — none of which Nexus uses on an OmniRig connection, as that same Settings page says. It now points at what actually matters: which OmniRig slot, and whether OmniRig's own window shows that radio online.

  • Special-event callsigns are recognised as CQs again, so double-click works on them.

    A call like II7MGBR or EN3SUKR does not fit the standard callsign shape, so FT8 sends it in a form that carries no grid — and Nexus was only treating the compound kind (the ones with a /) as real CQs. The rest were read as free text: no CQ chip, and double-clicking them did nothing, while the Work button in the Stations list started a QSO perfectly well. Both now behave the same way.

  • Your hidden panes and disabled modes survive an upgrade.

    If you had turned off modes you do not use, or hidden panels with the ⊞ menu, an upgrade could put them all back — those two choices were kept in browser storage rather than beside your settings, so anything that cleared it took them with it. They now live with your settings, per profile, and are included in the backup. A popped-out panel's own layout stays per-window, as it should.

  • Switching from CW to the FT screen left the rig on the CW frequency.

    FT8 came up on wherever CW had been — the mode changed to DIGU correctly, the dial did not. It only happened if you passed through Tempo on the way, which is why it looked intermittent: Tempo is a digital mode and asserts the rig mode, but it keeps its own band picker's frequency, and it was being counted as though it had already moved the dial for you. The FT screen then thought there was nothing to do.

  • A hardware CW keyer learns your speed before you send anything.

    With a WinKeyer, the speed slider only reached the keyer once you had sent something — so after launch the paddle ran at whatever speed the keyer itself was set to, and moving the slider did nothing until you typed a character. The keyer is now opened and told your speed as soon as it can key.

  • The filter width the radio actually took is now checked, not assumed.

    Some rigs accept a mode change with a filter width, answer "done", and quietly keep their own filter — so FT8 ended up on a 6 kHz SSB filter with nothing saying so. Nexus now reads the width back after a mode change and re-asserts it once if the radio kept its own, and if the second attempt is ignored too it tells you the actual width instead of implying success. Rigs that round to the nearest filter they own are left alone: asking for 3 kHz and getting 2.7 is the radio doing its job, not a fault.

  • A manual notch you can actually place, and a depth for the speech processor.

    The Notch button was driving the radio's automatic notch — the one that hunts a carrier down by itself — which is not what most operators mean by the word. The manual notch, the one you park on a whistle by ear, is now there too, with a frequency slider to put it where the whistle is. COMP gained the control it was missing: how hard the compressor works. Each appears only if your radio reports it, so nothing grows a slider with nothing behind it.

Added

  • The dial is marked on a native RF panadapter.

    If your radio streams its own spectrum (Icom CI-V, FlexRadio), the tuned frequency now has a line and a DIAL label on it. It is drawn only where the dial genuinely is: on a rig in FIXED scope mode, where the span is a band segment and the VFO sits wherever you tuned it, the line lands off-centre — and if the dial is outside the displayed window it is not drawn at all rather than pinned to the edge, which would say something untrue.

  • PSK and Tempo chapters in the manual.

    PSK31/QPSK31 shipped four releases ago with no chapter and Tempo never had one. Both now cover the tour, the workflows and the honest limits — and a test now fails the build if a shipped section has no chapter, so the next mode cannot reach a release undocumented.

  • AGC now offers AUTO and OFF, and stops reporting them as Mid.

    The Phone and CW cockpits gave three AGC settings — Fast, Mid, Slow — while the radios have five: an FT-710 offers AUTO and OFF on its own front panel. Worse than missing, they were misreported: a rig sitting on AUTO, which is where many operators leave it, displayed as Mid, and so did AGC switched off, because anything Nexus did not recognise folded to "mid". So the cockpit could state a setting the radio was not on. AUTO now appears to the left of Fast and OFF to the right of Slow, and a read-back of either shows what the rig is actually doing. Offered for every rig rather than guessed at: Hamlib does not report which AGC constants a backend accepts, so a rig that refuses one says so and the read-back shows what it really did.

Every version